Transcript Disclosure

Practical Tools for the Patient
Safety Officer:
Crafting Cultural Issues and
Understanding Trigger Tools
Frances A. Griffin, RRT, MPA
Director, Patient Safety
Institute for Healthcare Improvement
“Unsafe acts are like mosquitoes.
You can try to swat them one at a
time, but there will always be
others to take their place. The
only effective remedy is to drain
the swamps in which they breed.”
James Reason
Culture
A set of values, attitudes and
beliefs that governs behavior.
Culture is Context:
Human performance does not take
place in a vacuum – rather, it
takes place in an environment
engendered and maintained by:
 Management
 Governmental Regulators
 Front line personnel
From J. Bryan Sexton, PhD
Examples of Setting Culture
Organization vs. department / unit
 What do leaders talk about?
Teams
 Who is considered a member?
Orientation
 What do new staff hear?
On-going education
 How much and on what topics?
Errors & Adverse Events
How are they handled?
 System issue or individual blame?
 What is discussed and shared?
How do staff PERCEIVE they are
handled?
Impact of Culture
Turnover
Reporting
Practice
Service
Satisfaction
COST
A Safety Conscious Culture
Reporting
 Events, errors, unsafe conditions
Education
 All staff, new and on-going
Design
 Incorporation of human factors
Leadership
 Driving Force
Education & Training:
Key Questions
How many hours/year/employee?
How much is on patient safety?
What is the focus?
Does it include:
 Human factors awareness?
 Teamwork or CRM?
 Assertiveness or SBAR?
Orientation
Orientation
 Differences between formal &
informal
 Peer pressure
 Impact on turnover
Designing Systems for Safety
Prevention
 Design to prevent errors
Detection
 Make errors visible when they occur
Mitigation
 Reduce the harm when errors and
adverse events are not prevented or
detected
Designing for Safety
Reduce complexity
Optimize information processing
Automate wisely
Use constraints
Mitigate the unwanted side effects
of change
Thomas W. Nolan
High Reliability Organizations
Organizations that operate under
very trying conditions all the time
and yet manage to have fewer
than their fair share of accidents
“Managing the Unexpected”
Karl E. Weick & Kathleen M. Sutcliffe
“To the currently controversial question
of how many people die each year
from medical errors, the answers
range as high as the equivalent of two
fully loaded 747s crashing with no
survivors, each day of the year.
Hospitals aren’t even considered high
reliability organizations.”
Managing the Unexpected
Weick & Sutcliffe
Interventions to
Improve Culture:
Safety Briefings
Leadership WalkRounds
Human Factors Awareness
Training
SBAR Assertiveness Training
Crew Resource Management
Measuring Culture:
Safety Attitudes Questionnaire
J. Bryan Sexton, Ph.D.
The University of Texas at Center of
Excellence for Patient Safety and Practice
OR personnel report that briefings are important for patient safety, but not
common:
% of respondents who agree
100
100
90
80
70
86
81
92
89
60
50
40
30
20
32
22
10
22
27
24
0
Staff Surgeon
Surg Tech
Briefing Important
OR Nurse
CRNA
Anesthesiologist
Briefing Common
% of respondents within a clinical area reporting good teamwork climate
60
50
40
30
20
10
0
80
70
---------Post Briefings
---------PRE Briefings
100
Teamwork Climate Across Orgs
90
% of respondents within a clinical area reporting good safety climate
80
70
60
50
40
30
20
10
0
90
---------POST Briefings
---------PRE Briefings
100
Safety Climate Across Orgs
Improvements after a Cultural
Change
INCREASE: Nurse input is well received in the OR
INCREASE: I know the first and last names of all the
personnel that I worked with during my last shift
INCREASE: All OR personnel take responsibility for pt
safety
INCREASE: Pt safety is constantly reinforced as the
priority in the OR
INCREASE: Staffing levels are sufficient to handled the
number of patients
INCREASE: Personnel speak up if they perceive a
problem with pt care
DECREASE: High workload is common in the ORs here
Target: Safety Climate
Peter Pronovost, M.D., Ph.D., et al. at
Johns Hopkins
Administered Safety Climate Scale
before and after the intervention
Post intervention:
Marked improvement in Safety Climate at
each ICU
Reduced number of medication errors
Reduced LOS by 50%
Impact on ICU Length of Stay
Pronovost (2002)
2.5
ICU LOS
2
1.5
1
0.5
654 New Admissions: 7 Million Additional Revenue
May
Apri
l
Marc
h
Feb
Jan
Dec
Nov
Oct
t
Sep
Aug
ust
July
June
0
Key Points
Leadership Driven
 Must be visible
Slow to change
 Avoid “flavor of the month”
Fundamental to all safety
 Other initiatives will have limited
success
Lessons from other industries
 Aviation, nuclear power, etc.
Understanding Triggers
Why use Triggers?
Traditional reporting of errors,
incidents or events
 voluntary
 not reliable
• estimated at 10-20% of actual
 often involves violations of the 5 Rs
 includes errors that do not reach patient
In Search of Harm
Why is harm not reported?
 “known risk” or complication
 “cost of doing business”
Indicators
 Interventions
 Reversal agents
 Lab values
Background
Computerized triggers for ADE’s
 Brent James
ADE review identifying 14 triggers
 Samuel Henz
Idealized Design of the Medication
System – IHI & Premier
 modifications and testing
Preventability and Harm
Every system is designed to produce
the outcomes it gets
We have systems of care designed to
produce certain levels of harm
These levels of harm have become
acceptable as a property of the system
All harm is theoretically preventable
Definition of ADE
NCC MERP Index
A
B
C
D
E
F
G
H
I
Circumstances or events - capacity to cause error
Error occurred - did not reach the patient
Error reached patient, no harm
Monitoring or intervention , no harm
Temporary harm, intervention required
Temporary harm , initial or prolonged hospitalization
Permanent patient harm
Life sustaining intervention required
Death
Trigger Tool Advantages
Measures total harm
Moves from error but does not
exclude error
Easy with sampling over time
Measures accumulated efforts at
patient safety
Adverse Medication Events
New vs. Old
 Concentrates less
on errors
 Looks at all
unintended results
 Makes
measurement
easier
 Concentrates on
harm and those
errors that cause
harm
 Errors are the focus
of discussion
 Tends to focus only
on those results felt
to be related to error
 Requires judgement
 Human responsible
for most of the errors
Chart Review Triggers for ADE
 Diphenhydramine
 Vitamin K
 Romazicon
 Anitemetics
 Naloxone
 Antidiarrheals
 Kayexalate
 Serum glucose <50
 C. difficile positive
 PTT > 100 seconds
 INR >6
 WBC <3,000
 Platelet <50,000
 Digoxin level > 2
 Rising serum creatinine
 Oversedation / fall /
lethargy / hypotension
 Rash
 Abrupt medication stop
 Transfer to higher level
of care
Types of System Failures
Discrete Defect/Error
Poor Therapeutic Control
Information Retrieval and Processing
Predictable Risks including rare
extreme exacerbations of a known risk
Trigger Review Process
Random
Charts
Triggers
Reviewed
Pos
triggers
ID
Doses
Administered
No
End
Review
ADE’s/
1000 doses
Yes
Portion of
chart reviewed
ADE
Identified
No
Yes
End
Review
Harm
Category
Assigned
Determination of Harm
Was this preventable?
Is this the result of not doing
things right the first time?
Would I want this to happen to
me?
Multi-center Trigger Review
2837 charts reviewed using trigger tool
86 institutions
720 ADEs found on reviews
268,796 medications doses administered
ADE’s/1000 doses = 2.67
Admissions with ADE’s = 24.9%
Triggers Identifying ADEs
50%
45%
40%
35%
30%
25%
20%
15%
10%
5%
0%
43%
25%
8%
th
er
s
6%
O
PT
T>
1
00
8%
IN
R.
6
m
et
ic
An
tie
ve
rs
O
Ab
ru
pt
M
ed
St
ed
at
io
op
n
10%
Triggers in the ICU
Results from
Luther Midelfort











Positive blood culture
Abrupt drop in Hg >4gms
C. difficile positive
PTT > 100
INR > 6
Glucose < 50
Rising BUN +/or Serum
Creatinine to more 2x
baseline level
Radiologic test for emboli
or clot
Benadryl
Vitamin K
Flumazenil (Romazicon)












Naloxone (Narcan)
Antidiarrheals
Antiemetics
Sodium Polystyrene
(Kayexelate)
Code
Pneumonia onset in unit
Readmission to ICU
New onset dialysis
In unit procedure
Intubation / reintubation
Abrupt medication stop
Oversedation / lethargy /
hypotension
Adverse Events/ICU Day
Average .164 events/ICU Day
Range .04-.39 events/ICU Day
Luther Midelfort 2002
Data Results
 1294 total charts(Admissions) reviewed
 1450 events documented
 55% of admissions had adverse events
 28% of charts had more than 1 event
 18% related to medications
 11% coded on “E”codes
 8.9 day LOS with events
 4.3 day LOS without events
Luther Midelfort 2002
Top 10 Triggers
Trigger
# Positive
# With Harm
In Unit Procedure
628
112(17.8%)
Hct Drop
309
201(65%)
Intubation or Reintubation 309
166(54%)
Antiemetics
16(6.8%)
Luther Midelfort 2002
233
Top 10 Triggers
Trigger
# Positive # With Harm
PE Tests
200
35(17.5%)
Oversedation
184
159(86%)
Nos Pneumonia 158
154(97%)
Rising BUN
154
104(67%)
Pos Bld Culture 121
101(83%)
Med Stop
68(61%)
Luther Midelfort 2002
112
Events Related to Medications
Antibiotics 10%
Anticoagulants
24%
Electrolytes 2%
Insulin 8%
Luther Midelfort 2002
Narcotics 12%
Sedatives 24%
Other 17%
Consecutive Adverse Events













1-Iatrogenic pneumothorax
2-Sternal wound infection
3-Thrombophlebitis
4-Post Surgical bleed
5-ICU delirium
6-Nosocomial pneumonia
7-Theophyline
toxiciy/arrythmia
8-GI bleed
9-Iatrogenic pneumothorax
10-ICU delirium
11-Fluid overload
12-Oversedation
13-Urinary obstruction








14-ICU delirium
15-Rash
16-Aspiration pneumonia
17-Nausea
18-Pulmonary embolus
19-Nosocomial pneumonia
20-Sternal wound dehiscence
21-Dialysis induced
hypotension
 22-Severe hypotension with
NTG
 23-Renal failure post surger
 24-ICU delirium
 25-Sternal wound infection
Luther Midelfort 2002
Levels of Harm
 60 episodes event contributed to
death(4.1%)
 165 episodes event required intervention to
save life(11.4%)
 30 episodes event caused permanent
harm(2%)
 353 episodes event caused temporary harm
requiring hospitalization or prolonged
stay(24.3%)
 936 episodes event caused temporary harm
requiring intervention(64.5%)
Luther Midelfort 2002
Musings
NOI affect of events/admission $2739
1294 charts reviewed with 55% having
adverse events
710 charts had events X $2739
$2,000,000 affect on combined
collaborative NOIs
Local affect is about $2,000,000/year
Luther Midelfort 2002
Key Elements
Multidisciplinary team
 keep consistent
Review triggers only
 avoid “reading the chart”
Use data for internal comparison
 identify areas for further review
 drill down on specific triggers
Practical Process
For best results have 2 people review
each chart
Debrief after the 10 chart review
Reach an agreement on the events
Considerations
75% of all events will be picked up by
both reviewers
(these are the G,H,I harm levels)
25% of events will be picked up by one
or the other reviewer
(most often are E and F levels)
Definitions of harm become more
standard with 2 reviewers
Developing Triggers
Focus on:
 Type of event, location, population
List types of harm
Identify “clues”
Test with a team review
www.QualityHealthCare.org